Overview
Course material
We will use the book
- [ML] Machine Learning: A Bayesian and Optimization Perspective (ML), 2nd edition by Sergios Theodoridis, 2020. Download online from DTU Findit, or, the book can be purchased in polyteknisk bookstore at 10% discount.
As background material for the digital signal processing parts, we will use
Course outline by lecture module
| Week | Topic | Material (ML) |
|---|---|---|
| 1 | Digital signal processing, probability theory, machine learning | 1.1–2.3 |
| 2 | Matrix derivatives, constrained optimization, parameter estimation | 3.1–3.3, 3.5, 3.8–3.11, A.1–A.2, C.1–C.2 |
| 3 | Linear filtering | 2.4, 4.1–4.3, 4.5–4.7 |
| 4 | Adaptive filtering, LMS | 2.6, 5.1–5.5.1, 5.9, 5.12 |
| 5 | Adaptive filtering, RLS | 6.1–6.3, 6.5–6.8, 6.12 |
| 6 | Sparsity aware learning | 8.2, 8.10.1–8.10.2, 9.1–9.5, 9.9 |
| 7 | Shrinkage algorithms, Time-frequency analysis | 10.1–10.2, 10.5–10.6 |
| 8 | Dictionary learning, ICA, k-svd | 2.5, 19.1–19.3, 19.5–19.7 |
| 9 | Bayesian Modeling and EM | 11.2, 12.1–12.2, 12.4–12.5, 12.10 |
| 10 | State-space models, Hidden Markov models | 15.1–15.3.1, 15.7, 16.4–16.5 |
| 11 | State-space models, Kalman filter | 4.9–4.9.1, 4.10, 17.3 |
| 12 | Kernel methods, Kernel ridge regression | 11.1–11.5, 11.7 |
| 13 | Kernel methods, Support vector regression | 11.8 |